How do you handle model deprecations and version changes from providers?
30-second answerSay your answer out loud first, then reveal.
Process
- Inventory: a registry of model usage by feature, team, prompt version and volume.
- Watch announcements: subscribe to provider deprecation notices; calendar the retirement dates.
- Migration plan per feature:
• Run eval suites on the candidate replacement models.
• Tune prompts for the new model (behavioural differences: verbosity, tool-calling style, refusals).
• Compare cost and latency. - Rollout: shadow → canary → full; monitor quality, latency and cost.
- Don't wait for the deadline: start migrations early; deprecation deadlines cluster.
Aliases vs pinned versions: aliases (e.g. "model-latest") auto-update, which is convenient for prototypes but dangerous in production. Silent behaviour changes break workflows (see Q34).
Related
This is what real progress feels like.